A review of assistive technologies for people with Parkinson's disease.

The population is aging and with this, the incidence of age related diseases such as Alzheimer's disease (AD) and Parkinson's disease (PD) are increasing. Assistive Technology (AT) is viewed as one of the possible solutions which can be used to meet the needs of persons suffering from PD. AT can enable a person to carry out a task which otherwise they would be unable to undertake independently. An AT can have many functions which range from helping people to use a computer, to monitoring someone's condition. Within this paper we attempt to categorize the different types of AT for persons with PD. Each of the technologies will be compared and contrasted and an overview of what is currently available presented. The paper concludes with some visionary comments on how the current levels of AT may change in the future.

Review of ICT-based services for identified unmet needs in people with dementia.

Some of the needs that people with dementia and their informal carers currently perceive as insufficiently met by regular care and support services might be alleviated, or even be met, using modern Information and Communication Technology (ICT). The study described in this paper was designed to provide an insight into the state of the art in ICT solutions that could contribute to meet the most frequently mentioned unmet needs by people with dementia and their informal carers. These needs can be summarized as (1) the need for general and personalized information; (2) the need for support with regard to symptoms of dementia; (3) the need for social contact and company; and (4) the need for health monitoring and perceived safety. Databases that were searched include: PubMed, Cinahl, Psychinfo, Google (Scholar), INSPEC and IEEE. In total 22 websites and 46 publications were included that satisfied the following criteria: the article reports on people with dementia and/or their informal carers and discusses an ICT-device that has been tested within the target group and has proven to be helpful. Within the first need area 18 relevant websites and three studies were included; within the second need area 4 websites and 20 publications were included. Within the third and fourth need area 11 and 12 publications were included respectively. Most articles reported on uncontrolled studies. It is concluded that the informational websites offer helpful information for carers but seem less attuned to the person with dementia and do not offer personalized information. ICT solutions aimed at compensating for disabilities, such as memory problems and daily activities demonstrate that people with mild to moderate dementia are capable of handling simple electronic equipment and can benefit from it in terms of more confidence and enhanced positive effect. Instrumental ICT-support for coping with behavioral and psychological changes in dementia is relatively disregarded as yet, while support for social contact can be effectively realized through, for example, simplified (mobile) phones or videophones or (entertainment) robots. GPS technology and monitoring systems are proven to result in enhanced feelings of safety and less fear and anxiety. Though these results are promising, more controlled studies in which the developed ICT solutions are tested in real life situations are needed before implementing them in the care for people with dementia. It is recommended that future studies also focus on the integration of the current techniques and solutions.

Detection of workplace sedentary behavior using thermal sensors.

Sedentary behavior has been linked to leading causes of morbidity and mortality, including cancer, cardiovascular disease and diabetes. Those who work in office workplaces are susceptible to higher levels of sedentary behavior during the working day. This paper introduces a novel approach to the detection of sedentary behavior through the use of a thermal sensor mounted on the ceiling above a busy workspace. This solution was found to more accurately record 7 out of 10 activity metrics in comparison to self-assessment, when compared to chair pressure sensor recordings.

Evaluation of inherent performance of intelligent medical decision support systems: utilising neural networks as an example.

Researchers who design intelligent systems for medical decision support, are aware of the need for response to real clinical issues, in particular the need to address the specific ethical problems that the medical domain has in using black boxes. This means such intelligent systems have to be thoroughly evaluated, for acceptability. Attempts at compliance, however, are hampered by lack of guidelines. This paper addresses the issue of inherent performance evaluation, which researchers have addressed in part, but a Medline search, using neural networks as an example of intelligent systems, indicated that only about 12.5% evaluated inherent performance adequately. This paper aims to address this issue by concentrating on the possible evaluation methodology, giving a framework and specific suggestions for each type of classification problem. This should allow the developers of intelligent systems to produce evidence of a sufficiency of output performance evaluation.

An intelligent framework for the classification of the 12-lead ECG.

An intelligent framework has been proposed to classify an unknown 12-Lead electrocardiogram into one of a possible number of mutually exclusive and combined diagnostic classes. The framework segregates the classification problem into a number of bi-dimensional classification problems, requiring individual bi-group classifiers for each individual diagnostic class. The bi-group classifiers were generated employing Neural Networks (NN), combined with a combination framework containing an Evidential Reasoning framework to accommodate for any conflicting situations between the bi-group classifiers. A number of different feature selection techniques were investigated with the aim of generating the most appropriate input vector for the bi-group classifiers. It was found that by reducing the original input feature vector, the generalisation ability of the classifiers, when exposed to unseen data, was enhanced and subsequently this reduced the computational requirements of the network itself. The entire framework was compared with a conventional approach to NN classification and a rule based classification approach. The framework attained a significantly higher level of classification in comparison with the other methods; 80.0% compared with 66.7% for the rule based technique and 68.00% for the conventional neural approach.

Computerised electrocardiology employing bi-group neural networks.

A configuration of bi-group neural networks (BGNN) is proposed combined with an evidential reasoning framework to interpret 12-lead electrocardiograms for three mutually exclusive classes. A number of pre-processing feature selection techniques were investigated prior to application of the input feature vector to each individual BGNN. The network outputs were discounted within a belief interval of 1 based on their performance on test data prior to combination. It was found that the application of the feature selection techniques enhanced the individual performance of the BGNN, and subsequently enhanced the overall performance. The proposed framework was compared with conventional classification techniques of multi-output neural networks and linear multiple regression. The framework attained a higher level of classification in comparison with the other methods; 70.4% compared with 66.7% for both multi-output neural and statistical techniques.

Performance evaluation of artificial intelligence classifiers for the medical domain.

The application of artificial intelligence systems is still not widespread in the medical field, however there is an increasing necessity for these to handle the surfeit of information available. One drawback to their implementation is the lack of criteria or guidelines for the evaluation of these systems. This is the primary issue in their acceptability to clinicians, who require them for decision support and therefore need evidence that these systems meet the special safety-critical requirements of the domain. This paper shows evidence that the most prevalent form of intelligent system, neural networks, is generally not being evaluated rigorously regarding classification precision. A taxonomy of the types of evaluation tests that can be carried out, to gauge inherent performance of the outputs of intelligent systems has been assembled, and the results of this presented in a clear and concise form, which should be applicable to all intelligent classifiers for medicine.

Sensor-based activity recognition using extended belief rule-based inference methodology.

The recently developed extended belief rule-based inference methodology (RIMER+) recognizes the need of modeling different types of information and uncertainty that usually coexist in real environments. A home setting with sensors located in different rooms and on different appliances can be considered as a particularly relevant example of such an environment, which brings a range of challenges for sensor-based activity recognition. Although RIMER+ has been designed as a generic decision model that could be applied in a wide range of situations, this paper discusses how this methodology can be adapted to recognize human activities using binary sensors within smart environments. The evaluation of RIMER+ against other state-of-the-art classifiers in terms of accuracy, efficiency and applicability was found to be significantly relevant, specially in situations of input data incompleteness, and it demonstrates the potential of this methodology and underpins the basis to develop further research on the topic.

The creation of simulated activity datasets using a graphical intelligent environment simulation tool.

The availability of datasets capturing the performance of activities of daily living is limited by difficulties associated with the collection of such data. Software solutions can mitigate these limitations, providing researchers with the ability to rapidly generate simulated data. This paper describes the use of IE Sim to create a simulated intelligent environment within which activities of daily living can be performed using a virtual avatar. IE Sim has been demonstrated to facilitate the generation of datasets capturing normal activity performance in addition to overlapping activities and abnormal activities such as hazardous scenarios.

Pre-hospital body surface potential mapping improves early diagnosis of acute coronary artery occlusion in patients with ventricular fibrillation and cardiac arrest.

AIMS: To determine whether 80-lead body surface potential mapping (BSPM) improves detection of acute coronary artery occlusion in patients presenting with out-of-hospital cardiac arrest (OHCA) due to ventricular fibrillation (VF) and who survived to reach hospital. METHODS AND RESULTS: Of 645 consecutive patients with OHCA who were attended by the mobile coronary care unit, VF was the initial rhythm in 168 patients. Eighty patients survived initial resuscitation, 59 of these having had BSPM and 12-lead ECG post-return of spontaneous circulation (ROSC) and in 35 patients (age 69±13 yrs; 60% male) coronary angiography performed within 24 h post-ROSC. Of these, 26 (74%) patients had an acutely occluded coronary artery (TIMI flow grade [TFG] 0/1) at angiography. Twelve-lead ECG criteria showed ST-segment elevation (STE) myocardial infarction (STEMI) using Minnesota 9-2 criteria--sensitivity 19%, specificity 100%; ST-segment depression (STD) ≥0.05 mV in ≥2 contiguous leads--sensitivity 23%, specificity 89%; and, combination of STEMI or STD criteria--sensitivity 46%, specificity 100%. BSPM STE occurred in 23 (66%) patients. For the diagnosis of TFG 0/1 in a main coronary artery, BSPM STE had sensitivity 88% and specificity 100% (c-statistic 0.94), with STE occurring most commonly in either the posterior, right ventricular or high right anterior territories. CONCLUSION: Among OHCA patients presenting with VF and who survived resuscitation to reach hospital, post-resuscitation BSPM STE identifies acute coronary occlusion with sensitivity 88% and specificity 100% (c-statistic 0.94).

Flexible and customizable visualization of data generated within intelligent environments.

This paper outlines a tool for the visualization of data generated within Intelligent Environments. This tool has been designed with a focus on flexibility and customizability hence facilitating application to a range of areas including institutionalized or home-based healthcare monitoring. Through the use of an object toolbox, non-technical users can rapidly re-create a visual representation (aka a "Scene") of an intelligent environment and connect this scene to an active data repository. Data generated within the environment can be visualized in real-time, or summarized using a density ring visualization format that can be customized based on user defined rules to highlight events of particular interest. The tool was tested within a smart lab used as an active research environment. Collection of data over a one week period resulted in 3840 sensor activations. Visualization of this dataset illustrates the potential of the tool to highlight normal and abnormal activity trends within the environment.

The effects of electrode misplacement on clinicians' interpretation of the standard 12-lead electrocardiogram.

INTRODUCTION: This study investigates how a particular incorrect electrode configuration affects the 12-lead Electrocardiogram (ECG). METHODS: A correct and an incorrect 12-lead ECG were extracted from a 192-lead BSPM. This was done for 232 BSPMs yielding 464 12-lead ECGs. The particular incorrect ECG involved displacing electrodes V1 and V2 in the second intercostal space whilst also offsetting the remaining electrodes. These ECGs were examined in two stages: (a) analysis of the effects of electrode misplacement on signal morphology and (b) analysis of how often the incorrect electrode configuration changed the diagnosis of two clinicians in a random sample of 75 patients. RESULTS: According to the Root Mean Square Error (RMSE) of the difference between PQRST intervals in the correct and incorrect ECGs, lead V2 is the most affected lead (mean: 185 µV ± 82 µV), followed by lead V4 (mean: 114 µV ± 59 µV) and lead V1 (mean: 100 µV ± 47 µV). It was found that if the incorrect electrode configuration is applied, there is a 17% to a 24% chance the diagnostic interpretation will be different. Quantified using Similarity Coefficient (SC) leads V1 and V2 were found to be more alike when misplaced in the second intercostal space. The average SC between these leads when correctly placed was 0.08 (± 0.65), however when incorrectly placed, the average SC was 0.43 (± 0.3). CONCLUSION: There is a reasonable chance this particular incorrect electrode configuration will change the diagnosis of the 12-lead ECG. This highlights the importance of developing algorithms to detect electrode misplacement along with better education regarding ECG acquisition.

WiiPD--an approach for the objective home assessment of Parkinson's disease.

This paper introduces WiiPD, an approach to home-based objective assessment of Parkinson's disease. WiiPD aims to make use of the many capabilities of the Nintendo Wii Remote in combination with a number of bespoke data gathering methods to provide a rich and engaging user experience that can capture a wide range of motor and non-motor metrics. In this paper we discuss the architecture of the approach, and provide details of the implementation and testing of the motor-assessment component of the system. Initial results of testing on 6 users indicate that the system is able to differentiate between normal and abnormal motor performance, suggesting that the system has the potential to monitor the motor fluctuations associated with Parkinson's disease.

A web based tool for storing and visualising data generated within a smart home.

There is a growing need to re-assess the current approaches available to researchers for storing and managing heterogeneous data generated within a smart home environment. In our current work we have developed the homeML Application; a web based tool to support researchers engaged in the area of smart home research as they perform experiments. Within this paper the homeML Application is presented which includes the fundamental components of the homeML Repository and the homeML Toolkit. Results from a usability study conducted by 10 computer science researchers are presented; the initial results of which have been positive.

Towards a web-based knowledge management system for cardiovascular disease.

The objective of this paper is to examine the efficacy of the Knowledge Management (KM) paradigm for a web-based patient administration system (PAS) for cardiovascular disease (CVD). We discuss the role of contemporary information and communication technologies (ICTs) for the management of electrocardiographic information and how this can act as a foundation for a KM-based system.

Feature and classifier fusion for 12-lead ECG classification.

Two methodologies, feature and classifier fusion, for the problem of computerized 12-lead electrocardiogram classification, are investigated. Firstly, the entire classification problem is subdivided into a number of smaller bi-dimensional ones. By employing bi-group Neural Network classifiers, independent feature vectors for each diagnostic class are examined individually and the output from each classifier are fused together to produce one single result. Secondly, two classifiers, namely the aforementioned and a decision tree, are fused together through a novel approach of a Specificity Matrix. This methodology addresses the problem of unresolved conflict during fusion of classifiers and aims to exploit the merits of each classifier and suppress their weaknesses. 290 validated 12-lead electrocardiogram recordings, comprising six diagnostic classes, were used to train, validate and test both methodologies. The framework of bi-group classifiers enhanced the overall performance by 12.0% in comparison with conventional approaches. In the second instance, the fusion of the two classifiers produced a performance level of 81.3%; superior to either classifier in isolation. This approach offers a viable solution to the unresolved problem of conflict between classifiers during fusion and can be extended readily to accommodate any number of diagnostic classes and classifiers.

Implementation of intelligent decision support systems in health care.

The full implementation of any intelligent system in health care, which is designed for decision support, has several stages, from initial problem identification through development and, finally, cost-benefit analysis. Central to this is formal objectivist evaluation with its core component of inherent performance of the outputs from these systems. A Medline survey of one type of intelligent system is presented, which demonstrates that this issue is not being addressed adequately. Lack of criteria for dealing with the outputs from these "black box" systems to prescribe adequate levels of inherent performance may be preventing their being accepted by those in the health-care domain and, thus, their being applied widely in the field.